The science of ‘what-if?’

Pergamon Futures, Vol. 29, No. 6, pp. 533-539, 1997

0 1997 Elsevier Science Ltd. All rights reserved Printed in Great Britain

0016-3287/97 $17.00 + 0.00

PII: SO01 6-3287(97)00026-8

THE SCIENCE OF ‘WHAT-IF?’

Jerome R Ravetz

An appropriate response to the new challenges to science will require more

than new explicit goals and social organisation. New leading questions will be appropriate. The traditional questions of ‘what/how?’ for research and ‘how/why?’ for the design fields, will be supplemented by ‘what-if?‘. This pre-

viously had its place in exploratory phases of all research; now it will become an essential component of ‘post-normal’ science. Its form precludes the dog- matic and exclusive styles which have hitherto been dominant in science as applied to policy problems; and it lends itself to open enquiry and public partici-

pation. 0 1997 Elsevier Science Ltd

introduction: a new scientific style for new challenges to science

It is, by now, nearly universally acknowledged that the heart of the scientific enterprise

is no longer discovery, the triumphant advance of certain knowledge into the unknown. Only the spokesmen for ‘public understanding of science’ still remain unaware that the leading problems for science now derive from the challenges (and threats) presented by the hitherto blind and uncontrolled growth of our total scientific-technical-industrial system. Whether the issue be global carbon emissions, climate change, man-made diseases,

invasive alien substances such as xeno-estrogens, or even the social and cultural conse- quences of our runaway information-technology, both the problems and their appropriate styles of solution are new. They are radically different from those of the traditional curi- osity-motivated researcher, or even of the mission-oriented technology development team.

Let us consider the following statement:

‘With issues as complex as the impact of human activity on the natural environment, the search

for simple truths may obscure the uncertainty of reality.

Jerome R. Ravetz is Director of the Research Methods Consultancy, and can be contacted at 196 Clarence Gardens, London NW1 6AU, UK (Tel: + 44 (01171 224 7084; e-mail: 100105.251 [email protected].)

533

The science of ‘what-if?‘: J R Rave&

Almost anything we do, consume, or are exposed to has some risks. We have to decide which risks require tackling, with what priority, in what way, to what extent, and at what cost.

Environmental science is complex; describing technical data and detailed risk evaluation is difficult, but it reflects the reality of the choices society must make.

The issues are not simple and science matters.’

Those who are familiar with my critical writings will recognise some of the themes,

although perhaps the language is not familiar. The quotation is not from myself or some other environmental-activist science critic, but from Dr. Chris Fay, Chairman and Chief Executive of Shell UK Limited. It was made in May 1966, and is available on Shell’s

website named ‘brentspar”. Clearly, Dr. Fay received an education about the social problems of scientific knowledge which was not easily available in universities, but which was presented to him, rather unexpectedly, in the course of his work.

The quotation reminds us that the loss of certainty and the intrusion of ethics are central to this new syndrome of science. With it comes the loss of hegemony of scientific experts in the area of discussion and debate of science-related policy questions. A coher-

ent understanding of this new situation is only now developing. For it, we can draw on some elements of the critical tradition in studies of science that has developed over recent decades. The demystifications of Feyerabend, followed by the social-reductionist anthro-

pological studies, have made ‘objectivity’ a problem rather than a foundation for science. My own work was originally a mixture of radical social criticism of ‘big science’ with conservative nostalgia for ‘little science’. More recently, with Silvio Funtowicz, I have

placed traditional science in a framework that combines methodology with epistemology. We use the term ‘post-normal science’2, with its ironic reference to Kuhn’s seminal work,

to provide a historical location for this new sort of science. The concept of post-normal science rests on a three-fold distinction among different

sorts of problem-solving practices, based on the severity of either of the two attributes, systems uncertainties and decision stakes. The ‘inner’ zone, where both are low, we call

‘applied science’ (traditional ‘core’ or ‘basic’ science might be placed at the very corner of the diagram). When either is ‘medium’, we have ‘professional consultancy’, such as the practice of the surgeon or the consulting engineer. When either is severe, we are in the domain of ‘post-normal science’; and we argue that, in such circumstances, the quality-

assurance of the whole process requires an ‘extended peer community’ including all the relevant sorts of concerned lay persons. They bring with them their ‘extended facts’ based on their lived experience of the issue. Examples of post-normal science can be found in modern medicine, as with AIDS research-‘, and also public health in pollution and environmental issues. In the latter, the ‘extended facts’ provided by investigative journalism can be as crucial as the selected sample of the research results which is permitted

to enter the public domain. Up to now, Silvio Funtowicz and I have not systematically discussed the different

styles of enquiry appropriate to the different sorts of problem-solving; this is the topic that I explore here. For this I offer a threefold typology, in some ways paralleling that of post-normal science. By its means, we can see more clearly why new attitudes, giving credence to new sorts of questions, are necessary if science is to adapt to its new leading challenges.

A multiplicity of styles of research has long been recognised in science. At any time

534

The science of ‘what-if?‘: J R Ravetz

there is a hierarchy among disciplines; recently it has privileged those studying the more abstract and simple aspects of the external world. By the dominant criteria, data should be quantitative, and patterns of argument should be deductive and, if possible, formal.

Disciplines operating closer to our experienced reality tend to be considered as ‘soft’, and are forced into a state of ‘physics-envy’. The styles of investigation appropriate to their objects and functions (more intuitive and informal) are systematically squeezed out

by imitations of physics, which possess plausibility and effectiveness in varying degrees (frequently little). Every now and then, this awkward fact is admitted by a significant set of practitioners; and then the public learns of a degeneration within the field. Such now

seems to be the case in economics, where the style of abstract mathematical exercises is currently exposed for all as vacuous4. The causes of such pathologies are little studied; I discussed the problems of quality in my old book5. More recently, I have considered

the ways in which social aspects of scholarly disciplines can completely dominate their

criteria of quality, at the expense of their positive contenP.

Leading questions for different styles of science

Accepting the idea of different styles of research, with varying degrees of appropriateness, we can proceed to my classification, which is organized around ‘leading questions’. We

may start with three sorts of enquiry, which we may call research, design and exploration. For each of them, we have the relevant questions: respectively, ‘what/how?‘, ‘how/why?’ and ‘what-if?‘. This classification by questions is reminiscent of Aristotle’s four types of explanation, originally expressed like ‘the what’, ‘the how’, ‘the which’ and ‘the why’, later called the ‘material’, ‘efficient’, ‘formal’ and ‘final’ causes. Aristotle’s classification

was designed for his general theory of the acquisition of knowledge, with the principle of an analogy between ‘art’ and ‘nature’, so that every living thing is explained as if it

had been designed. My classification, by contrast, is organized around the different functions of a scientific enquiry, and so my categories will overlap with Aristotle’s.

These ‘leading questions’ are nothing like an absolute classification, nor should they be taken as defining the totality of any given sort of research. They are intended to illumi- nate rather than to define; and the term ‘leading question’ allows us to imagine that in

any particular inquiry, all three sorts of questions appear in their appropriate places. For research, the outcome of which is a statement intended to be factual, the leading question

is ‘what/how?‘. In Aristotle’s scheme, these questions deal with substance and agency: ‘what is this made of?‘, ‘how does this cause that?‘. Just as significant as the sorts of questions that are contained in ‘what/how’, are those which are excluded. These appear in the next leading question, ‘how/why?‘, were a ‘final cause’, a ‘function’ or even a

‘purpose’, is allowed. Such questions are applicable to the explanation of any artifact, a device or a tool. For what counts there, usually, is not so much its material substance or inner workings, but rather how its design enables it to perform its given function, to do its job. Of course there will be many overlaps between the two sorts of inquiry. Even

in research, instrumentation involves a design process, and so ‘how/why?’ is involved, though as a subsidiary rather than leading question in the research project; and in the sciences explaining structure (Aristotle’s ‘the why?‘) including much of biology, there is an inescapable rhetoric of design, in spite of the rigorous insistence on the absence of a Designer.

The leading question ‘what-if?’ is quite different in kind from the other two. in the

535

The science of ‘what-if?‘: J R Rave&

familiar sorts of scientific inquiry, it only appears in the earlier, exploratory phases of the

work, starting things off before problems are clearly defined and hypotheses hardened. In those contexts, ‘what-if?’ expresses the spirit of creativity, of inventiveness, of forays

into an unknown that is passive and expectant. In the context of the new challenges for science, ‘what-if?’ becomes the leading question, with a new urgency. Now our ignorance is no longer benign, but threatening. ‘what-if?’ a pathogenic agent in cattle is concentrated in infected feedstuffs and thereby made more virulent and transmissible? What if front-

loading gates are not checked as secure before an inherently unstable ferry gets underway? What if we continue to breed microorganisms that are resistant to ever more

of our antibiotics? What if artificial estrogens enter all animal food chains? Perhaps nothing, for some, most or all of the time; but perhaps something, just once, or once

too often. The use of ‘what-if?’ as a leading question is already enshrined in the practice of

risk assessment. In the case of hazards of chemical plants, the HAZOP methodology7 analyses unit actions in all their aspects: existence, space-time location and extent, as

well as quantity and intensity. Checklisting all the parameters, it asks ‘what-if’ any one lies outside the expected range. Then there might be an incident, an accident, or even a precursor to a disaster. By checklisting all the possibilities of ‘what-if?‘, the HAZOP methodology assures an effective management of this aspect, at least, of the concat- enation of unexpected, unanticipated, or unknown circumstances that eventually consti-

tute a disaster. Not all hazards can be so neatly broken down into known unit events; and in spite of HAZOP, major chemical plant accidents still occur. However, the recog- nition of ‘what-if?’ as a legitimate leading scientific question, along with ‘what/how?’ and

‘how/why?‘, would help us shape an effective scientific response to the new challenges and threats arising from our disturbance of the natural environment.

With ‘what-if?’ as the leading question, our whole conception of the scientific enterprise could evolve in a fruitful way. The prevailing attitude to uncertainty and com-

plexity would be transformed. In traditional science, as popularised and taught, uncertainty is an embarrassment. Normally, it is mentioned only when it has been completely tamed, as by the application of statistical routines. Until very recently, philosophers of science have searched for some simple methodology to explain the successes of science, smoothing over the real work of research in grappling with uncertainties of every sort. In the ‘how/why?’ research, oriented around ‘missions’ or devices performing a function,

uncertainty is respected, but only as something requiring to be managed. For in that case the task is to ensure that the uncertainties of the working environment of the system intrude only to an acceptable degree; and in much design and engineering practice,

uncertainties are already filtered through regulations or standard codes of practice, so that the problem-solving exercise deals mainly with these well-defined proxies for an uncertain reality. To enable a systematic confrontation of problem-solving practice with the world out there, we need the ‘what-if?’ approach.

Along with the uncertainty of the real world, we also need to deal effectively with its complexity. For centuries since Galileo and Descartes, our science has been built around the principle of studying the world in simplified, isolated bits. This approach pervades our whole scientific and technological culture; it is most easily noticed when it becomes challenged. Thus, we now begin to appreciate the limitations of a medical

practice which identifies disease with germs, and healing with germ-killing, thereby tend- ing to neglect both individual differences and the whole ecological and societal context

536

The science of ‘what-if?‘: / R Ravetz

of disease. Of course, successful routine practice of any sort requires simple tasks with

only moderate uncertainty. All we need is to appreciate that this ‘reductionism’ is a partial

view, and then to be open to the importance of perspectives that are complementary to

it. indeed, it is when we are accustomed to asking ‘what-if?‘, and, as it were, expecting

the unexpected, that we fully appreciate how no single perspective can completely cap- ture any real situation. This is what genuine complexity, as opposed to mere compli-

cation, is all about. The adoption of ‘what-if?’ as a scientific style will also have important effects on the

policy process. We now know that the application of science to policy is very different from a simple process of ‘getting the facts’. For the facts are not merely ‘got’; to secure

them requires an initial setting of priorities so that research is supported; and then the research must be designed around appropriate questions; for every policy issue is com-

plex, including aspects of both nature and society, and where the ‘cause’ of the problem will be equally complex. The history of the BSE disaster shows how research that is too

little, too late, and focused away from the crucial policy questions, can allow a crisis to mature over the years to the point of no return, in spite of the best efforts of sincere and

dedicated scientists.

The style of ‘what-if?’ is also an expression of the Precautionary Principle, which is becoming increasingly important in environmental policy. It protects scientists against a

premature choice of research problems to be investigated, and a hasty closing down of

exploration. One could imagine a continuous dialogue among the participants of the different sorts of inquiry. Although ‘what-if?’ necessarily moves from centre stage once

the research effort gets underway, it should always be fostered as an essential complement to the puzzle-solving from which the relevant ‘facts’ eventually emerge.

There are good reasons why ‘what-if?’ science should emerge just now. The question of ‘safety’ is emerging as a great task for science, as a successor to the twin goals of

knowledge and power that were articulated nearly four centuries ago in the Scientific Revolution. We find ourselves in a somewhat paradoxical situation: on the one hand,

people in many different societies have never been so safe; and on the other hand there

has never been so great a public consciousness of danger. In some ways, the paradox is in our minds, for as we come to believe that (thanks to science) our lives can be truly safe, we demand the elimination of any danger that we encounter (and frequently blame

science for its presence). However, there is something real out there as well; as the dangers have diminished, their character has changed. They are no longer typified by the

devastating plagues and widespread natural disasters of the past. They have tended to become more subtle, perhaps more insidious, even menacing. We experience threats which are totally novel, such as genetic damage from nuclear radiation that will extend indefinitely into the future; and we confront new diseases whose origins are as uncertain and complex as their treatments. In the face of such novel sorts of dangers, there is no effective alternative to a ‘what-if?’ style of science. The complex condition of safety will

not be achieved by the simplistic, reductionist tradition of science embodied in the ‘what/how’ and ‘how/why?’ styles. We need ‘what-if?’ as a complement to them, to be developed and employed in the appropriate contexts in the appropriate ways.

Finally, there are the social implications of the ‘what-if?’ style of science. The puzzle- solving exercises of ‘normal science’, or indeed of any routine practice, require and foster a ‘convergent’ style, in which a standardized expertise is necessary and dominant. The ‘what-if?’ style operates on the margins of that sort of expertise. Insights and suggestions

537

The science of ‘what-if?‘: 1 R Ravetz

from people with different sorts of expertise or even none at all, must be accepted into the dialogue. In connection with post-normal science we have described this as an

‘extended peer community’. There we introduced the idea in connection quality-assurance of the scientific materials introduced into the policy process. The ‘what-if?’ style

performs another function for these extended peer communities; and even more strongly it justifies their ‘extended facts’. These are data which derive from sources outside ortho-

dox research; they may come from anecdotes, traditional knowledge, informal studies, or investigative journalism. Since so many of the ‘what-if?’ questions are prompted by such unorthodox data, the adoption of the ‘what-if?’ style is inseparable from the recog-

nition of extended facts, extended peer communities, and the approach of post-normal science.

Conclusion: the contexts of ‘what-if?’

In the context of a collection of essays on ‘Visions of Science’, this exploration of a

methodological idea might seem rather pale and tame. What about the ‘critical science’ of a generation ago?a There are two related reasons for my setting aside such a politicised

approach, at least for the moment. One is the state of the politics of science at the present time. For a while, when the reality of the ‘military-industrial-scientific complex’ was a fresh discovery for a broad public, in connection with The Bomb and Vietnam, it seemed possible that a radical movement could be organized around science itself. The American

‘Science for the People’ and the British ‘Society for Social Responsibility in Science’ were optimistic responses to the militant mood of the later 1960s.

However, in the short run at least, that enthusiasm could not be sustained. The focus

of protest shifted to ‘the environment’, in which science itself was not seen as problem-

atic; and the really corrosive criticism of science, demystifying its claims to objective knowledge, was conducted ostensibly as an academic exercise, by the post-Feyerabend

critical schools. In the meantime, politics itself has changed. The collapse of Socialism as a viable alternative social system has gravely weakened the radical forces on the traditional issues of distribution. The ‘leading contradiction’ of the industrial system has

shifted to that of ‘sustainability’ or, in other words, survival. If science is to become a

central issue within that new context, debate will focus less on distributional aspects (access to jobs, and utilisation of results), and more on its philosophical foundations.

This new focus is expressed in Zia ‘s campaign for ‘Others’9. As it matures, this should provide us with the precious gift of ‘seeing ourselves as others see us’. On that basis, we might accomplish the necessary reform of science, in its social functions, work-

ing methods, and conceptual objects. This change could be as profound as that which took place four hundred years ago, in the Scientific Revolution. However, this new reform is, as yet, in a very early stage, partly because the cultures which are candidates for leading ‘Other’ status still lack the coherence and self-confidence that would enable them to engage in a dialogue. It is likely that each of them will need to go through a phase of rejection and hostility (in the religious sphere manifesting as ‘fundamentalism’) before any cross-cultural dialogue, however critical, could emerge.

A complementary development could now be taking place within occidental culture, among the tendencies variously labelled ‘green’, ‘feminist’, or ‘New Age’. The variety of

names indicates the great variety of issues and styles, some overlapping but some mutu- ally incompatible and hostile. Any consistent vision of science based on one section of

538

The science of ‘what-if?‘: J R Ravetz

this diverse movement would certainly alienate most of the other sections in some way.

However, this movement, based on a widespread disillusion with one or another preten-

sion of our modern industrial system, is still in an early stage of growth. Scarcely a gener-

ation separates us from the ‘counter-culture’ of the 1960s when it all exploded. For the moment, then, my approach has been to develop ideas on particular pieces

of the jigsaw. these include my previous studies on uncertainty and post-normal science conducted with Silvio Funtowicz, and those done with or for Zia, as were published in The Merger of Know/edge with PowerlO. None of those explorations has yet had (to my

knowledge) a resonance anywhere in the occident. From this, I infer that any synthesis I attempted now would be either far ahead of its time or off the mainstream of develop-

ment. So I prefer not to rush the work, but instead to cultivate my understanding on

various special problems. The ‘what-if?’ style of science is one of these; at the same time I am working with Silvio Funtowicz on the,theory of ‘reflexive complex systems’. After

that should come a study of dialectics (as I understand the idea), which should deepen the theories of knowledge and of practice which I first articulated in Scientific Knowledge

and its Social Problems’ ’ . This has been a lengthy apology for the essays that I chose not to write for this

present collection. I hope that it puts ‘the science of what-if?’ in perspective both as a

contribution to a vision of science, and as significant in its own right.

Notes and references

1. 2. 3.

4. 5.

6.

7.

8.

9.

10. 11.

httpd/www/sheIlexpro.brentspar.com S. Funtowicz and J.R. Ravetz, Science for the Post-Normal age. futures 25(7), 735-755 (1993). Epstem, S., The construction of lay expertise: Aids activism and the forging of credibility in the reform of clinical trials. Science Technology and Human Values, 1995, 20(4), 408437. J. Cassidy, The end of economics. The New Yorker 2 December, 50-60 (1996). Ravetz, J. R., Scientific Know/edge and its Social Problems (with a new introduction). Transaction Press, New Brunswick, NJ, 1996. Ravetz, J.R., Economics as an elite folk-science: the suppression of uncertainty. The Journal of Post-Keyne- sian Economics, 1994/5, 17(2), 165-l 84. Chemical Industries Association, Hazard and Operability Studies. Chemical Industries Association, Lon- don, 1992. Ravetz, J. R., Critical Science: politics and philosophy. In Scientific Know/edge and its Social Problems. Transaction Press, New Brunswick, NJ, 1996, pp. 422-436. Z. Sardar, Conquests, chaos, complexity. The Other in modern and post-modern science. Futures, 1994, 26(6), 665-682. Ravetz, J. R., The Merger of Knowledge with Power. Cassell, London, 1990. Ravetz, J.R., Scientific Knowledge and its Social Problems (with a new introduction). Transaction Press, New Brunswick, NJ, 1996.

539

Documents

The science of ‘what-if?’